novel-master
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a standard creative writing assistant. No malicious instructions, obfuscation, or unauthorized resource access were detected across the analyzed files.
- [PROMPT_INJECTION]: The skill uses clear, instruction-based language focused on its role as a writer and editor. There are no attempts to bypass safety filters or reveal underlying system prompts.
- [DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or unauthorized data transfer patterns are present. Network operations are restricted to searching for public social media trends and literary elements.
- [REMOTE_CODE_EXECUTION]: The skill contains no code execution patterns, package installations, or script downloaders.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from web searches for novel trends.
- Ingestion points: Tool outputs from
web_search,WebSearch, or browser tools (referencing current trends and hot topics). - Boundary markers: Not explicitly defined in instructions, though the skill uses structured markdown templates for all outputs.
- Capability inventory: Limited to creative text generation and web search tools for research. No file system writes or subprocess execution capabilities were identified.
- Sanitization: The skill uses specific search keyword strategies to limit the scope of ingested data to literary and trend-related topics.
Audit Metadata